Focal neocortical epilepsy is a largely intractable medical problem, with most cases responding poorly to anti-convulsive medications and current surgical treatment options limited because of the likelihood of neurological deficits. Because much information processing is vertically organized in cortex, while seizure propagation occurs primarily through lateral connections, a series of incisions could prevent the spread or initiation of seizures but largely preserve function. These incisions must not cut the blood vessels on the brain surface and it remains unclear which cortical layers are best cut to achieve optimal seizure control and minimal neurological impact. Tightly-focused femtosecond laser pulses provide a unique tool to make micrometer-scale cuts several millimeters within the bulk of a tissue with minimal collateral damage. We hypothesize that using these cuts to transect the neural connections in targeted cortical layers will block the initiation or propagation of focally initiatd epileptic seizures. Because these cuts target only horizontal connections in a specific cortical layer and the majority of the neural connectivity of the cortex is preserved, there will be minimal neurological deficit. A primary goal of this proposal is to determine which cortical layer(s) shoul be cut and in what geometric pattern to maximally interfere with seizure initiation and propagation while minimally affecting normal function. In Aim 1, we test the acute efficacy of femtosecond laser cuts to specific cortical layers in preventing epilepsy initiation and propagation. Epileptic seizures are modeled in rats by microinjection of 4-aminopyridine into cortex. Local field potential recordings and two-photon calcium sensitive dye imaging are used to monitor neural activity and seizure propagation. Femtosecond laser ablation is used to encircle or subdivide the seizure initiation site with subsurface cuts. First, we determine which cortical layers must be transected to prevent seizures from propagating outside of the encircled region. The goal is to determine the minimum number of layers to cut for seizure containment. Building on recent data that suggests that clinical seizures result from the coalescence of microseizures, we next investigate whether a grid pattern incised at the seizure initiation site can prevent seizure initiation by separating microdomains. In Aim 2, we explore potential side effects of the most promising laser cuts from Aim 1 by recording changes in evoked signals in somatosensory cortex after whisker stimulation. These experiments will test, in animal models, a new laser-based surgical method for the treatment of focal neocortical epilepsy. In addition, this work will provide valuable, in vivo data on cortical layer- specific initiation and propagatio of seizures. If the acute animal model experiments proposed here as well as future studies that evaluate the longer-term effectiveness prove successful then human implementation is feasible using recently-developed laser technology and would enable layer-specific cuts to be produced in all but the bottom of sulci, opening the door to new surgical treatments for epilepsy.